Our long term research goal is to understand mechanisms for regulation of the receptor-coupled adenylyl cyclase signaling pathway. We have been using the beta-adrenergic receptor (BAR) as a model and have focused on differences in regulation of BAR subtypes. Two areas are addressed in this year's annual report: A) Differential Sorting of Human Beta-1- and -2-Adrenergic Receptors after Clathrin-Mediated Endocytosis. There are three subtypes of the human beta-adrenergic receptor, B1AR, B2AR and B3AR, which we and others have shown to differ in their ability to be regulated (B2AR>B1AR>>B3AR). Thus, B1AR undergoes less agonist-mediated internalization than B2AR. Previously, we showed that: (i) overexpressing either G protein-coupled receptor kinases or arrestins enhanced internalization of B1AR to that of B2AR; (ii) B1AR was endocytosed by the clathrin-coated pit pathway in an arrestin- and dynamin-dependent manner similar to that of B2AR; (iii) and using a green fluorescent protein (GFP)-tagged B1AR to visualize endocytosis, B1AR-GFP upon agonist exposure redistributed from the plasma membrane to endosomes, some containing fluorescent-labeled transferrin, an endosomal marker. Although B1AR and B2AR use the same endocytic pathway, we recently found that soon after endocytosis, the two subtypes were sorted differently. For these studies, we used baby hamster kidney (BHK) and human embryo kidney (HEK 293) cells transfected with either wild-type BARs or B1AR-GFP and hemagglutinin-tagged B2AR. For most experiments, the cells were transfected with arrestin-2 to ensure equal and extensive agonist-mediated internalization of both subtypes. Whereas B1AR remained in endocytic vesicles close to the cell surface, B2AR was found in endosomes deeper in the cytoplasm. Furthermore, although binding assays showed that both subtypes were internalized by 50-60%, more B1AR remained associated with the plasma membrane than B2AR after agonist stimulation. Using subcellular fractionation by sucrose density gradient centrifugation, we confirmed the difference in subtype sorting. In agonist-treated cells, 50% or more of B2AR redistributed from the more dense plasma membrane fraction to a less dense endosomal fraction. In contrast, less than 25% of B1AR appeared in an endosomal fraction that had a slightly higher apparent density than those containing B2AR. When the gradients were centrifuged for 90 instead of 60 min, the two populations of endosomes appeared at the same density. Such behavior indicates differences in their properties. Thus, it appears that shortly after the two subtypes are internalized through the clathrin-mediated pathway, they are segregated to different endosomal compartments. This differential sorting may result in a number of physiological consequences. In this regard, last year we reported that when expressed in the same cells used in our current studies, B2AR but not B1AR underwent agonist-mediated down-regulation. It is believed that down-regulation occurs when the receptors are sorted to late endosomes and eventually to lysosomes. B1AR may be in endosomes that do not converge with the degradative pathway. B) Targeting BAR Subtypes to Caveolae: Differences between Endogenously and Heterologously Expressed B1AR. In recent studies, different conclusions were reached about the localization of BARs in caveolae, microdomains of the plasma membrane that are cholesterol- and sphingolipid-enriched. Some reported that both B1AR and B2AR are restricted to caveolae along with adenylyl cyclase and other signaling proteins, whereas others found that only B2AR is so targeted and B1AR is distributed more randomly. To resolve this issue, we examined the localization of the two subtypes either separately expressed in stably transfected BHK and HEK 293 cells; endogenously co-expressed in rat C6 glioma cells; or separately expressed and inducible in C6 cells. Caveolae-enriched fractions were purified from each cell line by a detergent-free method. Briefly, cells were scraped into buffered sodium carbonate, homogenized, adjusted to 45% sucrose, layered under a discontinuous 5-35% sucrose gradient and centrifuged at 150,000 x g for 16-18 h. Gradient fractions were collected, assayed for BAR binding activity and protein, and analyzed by immunoblotting with antibodies to BARs and caveolins, proteins that are markers for caveolae. Both human subtypes co-localized exclusively with caveolin at the 5/35% sucrose interface when expressed at ~1-2 pmol/mg protein in BHK and HEK 293 cells. Fractionation of rat C6 cells revealed that while endogenous B2AR (~40 fmol/mg) also was restricted to the caveolin-containing fractions, endogenous B1AR (~80 fmol/mg) was more randomly distributed. When rat B1AR was induced in C6 cells (0.6-1.5 pmol/mg), however, it was targeted to caveolae as was the induced rat B2AR. Both subtypes, whether endogenous or induced, functioned similarly in mediating agonist-stimulated cAMP accumulation in C6 cells. Thus, there appears to be a thresh-hold effect in targeting of B1AR but not B2AR to caveolae.